- MMLU-Pro 86.7
- MMLU-Redux 94
- C-Eval 91.9
- SuperGPQA 67.1
- IFEval 93.4
- IFBench 76.1
- MultiChallenge 61.5
- AA-LCR 66.9
- LongBench v2 60.2
- Best of HLE w/ CoT 25.3 / HLE w/ tool 47.5 47.5
- GPQA Diamond 86.6
- HMMT Feb 25 91.4
- HMMT Nov 25 90.3
- SWE-bench Verified 72
- Terminal Bench 2 49.4
- LiveCodeBench v6 78.9
- CodeForces (Qwen query set) 2100
- OJBench 39.5
- Mean of FullStackBench en 62.6 / FullStackBench zh 58.7 60.7
- BFCL-V4 72.2
- TAU2-Bench (airline fixes from Claude Opus 4.5 system card) 79.5
- VITA-Bench 33.6
- DeepPlanning 24.1
- Browsecomp (context folding) 63.8
- Browsecomp-zh 69.9
- WideSearch 60.5
- Seal-0 44.1
- MMMLU 86.7
- MMLU-ProX (29 languages) 82.2
- NOVA-63 58.6
- INCLUDE 82.8
- Global PIQA 88.4
- PolyMATH 68.9
- WMT24++ (XCOMET-XXL, 55 languages) 78.3
- MAXIFE 87.9
- MMMU 83.9
- MMMU-Pro 76.9
- MathVision 86.2
- Mathvista(mini) 87.4
- DynaMath 85.9
- Best of ZEROBench 9 / ZEROBench_sub 36.2 36.2
- VlmsAreBlind 96.7
- BabyVision 40.2 / 34.5 (with CI / without CI) 40.2
- RealWorldQA 85.1
- MMStar 82.9
- MMBenchEN-DEV-v1.1 92.8
- SimpleVQA 61.7
- HallusionBench 67.6
- OmniDocBench1.5 89.8
- CharXiv(RQ) 77.2
- MMLongBench-Doc 59
- CC-OCR 81.8
- AI2D_TEST 93.3
- OCRBench 92.1
- ERQA 62
- CountBench 97
- RefCOCO(avg) 91.3
- ODInW13 44.5
- EmbSpatialBench 83.9
- RefSpatialBench 69.3
- LingoQA 80.8
- Hypersim 12.7
- SUNRGBD 36.2
- Nuscene 15.4
- Best of VideoMME(w sub.) 87.3 / VideoMME(w/o sub.) 83.9 87.3
- VideoMMMU 82
- MLVU 87.3
- MVBench 76.6
- LVBench 74.4
- MMVU 74.7
- ScreenSpot Pro 70.4
- OSWorld-Verified 58
- AndroidWorld 66.4
- TIR-Bench 53.2 / 42.5 (with CI / without CI) 53.2
- V* 93.2 / 90.1 (with CI / without CI) 93.2
- SLAKE 81.6
- PMC-VQA 63.3
- MedXpertQA-MM 67.3
Qwen3.5-122B-A10B
Qwen3.5-122B-A10B places in the top three on 5 benchmarks.
Qwen Released 24 Feb 2026 94
Benchmarks with a score
0
Ranked first, out of five or more models
5
In the top three
Qwen Feb 2026 Qwen3.5-122B-A10B
Open weights
No Epoch Capabilities Index score yet
- IFBench 76.1% #19 /70
- MMMU-Pro 76.9% #25 /72
- AA-LCR (Artificial Analysis Long Context Reasoning) 66.9% #13 /31
Where it ranks
Its best score on each benchmark, and where that ranks among every model with a score there. We rank each model once, at its best setting.
Knowledge and language
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| MMLU-Pro | 86.7% | #13 of 151 | Model card | |
| IFEval | 93.4% | #11 of 126 | Model card | |
| C-Eval | 91.9% | #7 of 38 | Model card | |
| MMLU-Redux | 94% | #7 of 38 | Model card | |
| SEA-HELM | 74.7 | #15 of 62 | Official leaderboard | |
| HallusionBench | 67.6% | #6 of 23 | Model card | |
| IFBench | 76.1% | #19 of 70 | Model card | |
| Humanity's Last Exam | 47.5% | #46 of 147 | Model card | |
| LMArena (Elo) | 1,416 | #108 of 333 | Official leaderboard | |
| Judgemark v2.1 | 81.0 | #14 of 42 | Official leaderboard | |
| MMMLU | 86.7% | #21 of 62 | Model card | |
| SuperGPQA | 67.1% | #9 of 26 | Model card | |
| MMLU-ProX | 82.2% | #6 of 17 | Model card | |
| INCLUDE | 82.8% | #9 of 25 | Model card | |
| AA-LCR (Artificial Analysis Long Context Reasoning) | 66.9% | #13 of 31 | Model card | |
| MultiChallenge | 61.5% | #17 of 36 | Model card | |
| LongBench v2 | 60.2% | #13 of 27 | Model card | |
| Longform Creative Writing (EQ-Bench) | 54.3 | #78 of 119 | Official leaderboard | |
| Extended NYT Connections | 51.7% | #65 of 97 | Official leaderboard | |
| Vectara Hallucination Leaderboard (2025 dataset) | 11.2% | #73 of 104 | Official leaderboard | |
| SimpleVQA | 61.7% | #17 of 23 | Model card | |
| NOVA-63 | 58.6% | #4 of 5 | Model card | |
| WideSearch | 60.5% | #16 of 17 | Model card | |
| Global PIQA | 88.4% | #8 of 8 | Model card | |
| MAXIFE | 87.9% | #5 of 5 | Model card | |
| Seal-0 | 44.1% | #5 of 5 | Model card |
Math and science
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| GPQA Diamond | 86.6% | #59 of 284 | Model card | |
| PolyMath | 68.9% | #6 of 18 | Model card | |
| DTBench | 73.8% | #71 of 178 | Official leaderboard | |
| HMMT 2025 | 91.4% | #20 of 49 | Model card | |
| CritPt | 0.86% | #97 of 160 | Epoch AI | |
| HMMT Nov 2025 | 90.3% | #14 of 23 | Model card |
Coding
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| LiveCodeBench v6 | 78.9% | #32 of 76 | Model card | |
| SWE-bench Verified | 72% | #65 of 150 | Model card | |
| Codeforces rating | 2,100 | #18 of 41 | Model card | |
| OJBench | 39.5% | #7 of 12 | Model card | |
| FullStackBench | 60.7% | #4 of 6 | Model card | |
| Code Arena | 1,360 | #93 of 124 | Official leaderboard |
Agents and tools
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| SnakeBench | 33 | #22 of 183 | Official leaderboard | |
| BFCL V4 | 72.2% | #5 of 24 | Model card | |
| BrowseComp-ZH | 69.9% | #7 of 17 | Model card | |
| ScreenSpot-Pro | 70.4% | #12 of 25 | Model card | |
| tau2-bench | 79.5% | #37 of 69 | Model card | |
| LisanBench | 392.7 | #73 of 133 | Official leaderboard | |
| AndroidWorld | 66.4% | #11 of 18 | Model card | |
| Terminal-Bench 2.0 | 49.4% | #39 of 61 | Model card | |
| VITA-Bench | 33.6% | #8 of 12 | Model card | |
| BrowseComp | 63.8% | #52 of 77 | Model card | |
| Mystery Game Puzzles | 17% | #56 of 80 | Epoch AI | |
| MobileWorld | 35% | #15 of 18 | Official leaderboard | |
| OSWorld-Verified | 58% | #32 of 37 | Model card | |
| DeepPlanning | 24.1% | #6 of 6 | Model card |
Professional work
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| MedXpertQA-MM | 67.3% | #6 of 12 | Model card | |
| PMC-VQA | 63.3% | #3 of 4 | Model card | |
| SLAKE | 81.6% | #3 of 4 | Model card |
Vision, audio and robotics
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| MMMU | 83.9% | #7 of 97 | Model card | |
| MathVista | 87.4% | #6 of 64 | Model card | |
| MMBench-EN v1.1 | 92.8% | #2 of 17 | Model card | |
| OCRBench | 92.1% | #4 of 28 | Model card | |
| MMStar | 82.9% | #5 of 30 | Model card | |
| DynaMath | 85.9% | #3 of 17 | Model card | |
| MLVU | 87.3% | #3 of 16 | Model card | |
| CountBench | 97% | #4 of 20 | Model card | |
| ZeroBench | 36.2% | #6 of 25 | Model card | |
| MVBench | 76.6% | #4 of 15 | Model card | |
| AI2D | 93.3% | #12 of 44 | Model card | |
| MATH-Vision (MathVision) | 86.2% | #13 of 42 | Model card | |
| Video-MME | 87.3% | #13 of 42 | Model card | |
| RealWorldQA | 85.1% | #10 of 30 | Model card | |
| Hypersim | 12.7 | #1 of 3 | Model card | |
| MMMU-Pro | 76.9% | #25 of 72 | Model card | |
| VLMs Are Blind | 96.7% | #3 of 8 | Model card | |
| MMLongBench-Doc | 59% | #5 of 13 | Model card | |
| VideoMME v2 (with subtitles) | 26.7% | #12 of 31 | Official leaderboard | |
| MMVU | 74.7% | #8 of 20 | Model card | |
| Vision Arena | 1,226 | #63 of 136 | Official leaderboard | |
| V* | 93.2% | #6 of 12 | Model card | |
| RefCOCO / RefCOCO+ / RefCOCOg | 91.3% | #10 of 19 | Model card | |
| LVBench | 74.4% | #14 of 25 | Model card | |
| Video-MMMU | 82% | #18 of 30 | Model card | |
| EmbSpatialBench | 83.9% | #6 of 10 | Model card | |
| SUNRGBD | 36.2 | #3 of 5 | Model card | |
| CharXiv | 77.2% | #41 of 67 | Model card | |
| ERQA | 62% | #16 of 26 | Model card | |
| CC-OCR | 81.8% | #5 of 8 | Model card | |
| RefSpatial-Bench | 69.3% | #5 of 8 | Model card | |
| ODinW-13 | 44.5% | #6 of 9 | Model card | |
| BabyVision | 40.2% | #16 of 20 | Model card | |
| LingoQA | 80.8% | #4 of 4 | Model card | |
| Nuscene | 15.4 | #2 of 2 | Model card | |
| TIR-Bench | 53.2% | #2 of 2 | Model card |
Other
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| LMCA | 32.2% | #77 of 139 | Epoch AI | |
| Conceptual Reasoning Index | 45.7 | #83 of 149 | Official leaderboard | |
| ACCoRD | 58.0 | #94 of 150 | Official leaderboard |
Its launch post
The benchmarks the lab chose to headline, with the scores it gave.